WANI Dictionary

LASSO regression

The LASSO (least absolute shrinkage and selection operator) regression is one of the regularized linear regression methods. It can also be applied to logistic regression [2].

The regularized regression applies a penalty to prevent multicollinearity and overfitting issues. The LASSO regression also selects independent factors to include in the regression model and is thus frequently used to handle high dimensional data.

Definitions in the literature

• The advantage of LASSO regression is that we consider all potential drivers but only a subset of the covariates are selected [1].